Real-time data warehousing for business intelligence

Farooq, Farrah; Sarwar, Syed Mansoor

doi:10.1145/1943628.1943666

Cited by 9 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additional materialized cubes add extra overhead to the incremental loading phase. On the contrary, semi-structured MCMs have more capability to incorporate real-time data changes (Farooq & Sarwar, 2010).…”

Section: Storage Estimations and Incremental Loading Of Data Updatesmentioning

confidence: 99%

“…Generally, there are three types of ETL jobs: initial‐loading ETL job to first time load the ODS data to DW repository, full‐reloading ETL job to refresh the DW, and incremental‐loading ETL jobs to incorporate data updates. In (Farooq & Sarwar, ), we discuss the implementation of sample RTDW architecture and evaluate three types of ETL jobs on relational and XML MCMs. Our experimental results show that by using incremental ETL jobs, the semi‐structured XML model has more capability for incorporating real‐time updates from operational sources than structured relational MCM.…”

Section: Dwvf and Prototype Implementationmentioning

confidence: 99%

See 1 more Smart Citation

The data warehouse virtualization framework for operational business intelligence

Farooq

2012

Expert Systems

Self Cite

View full text Add to dashboard Cite

In order to explore the most current information and react faster to changing business conditions, organizations consider real‐time data warehousing a powerful technique to achieve operational business intelligence (BI). We propose in this paper a novel real‐time data warehouse (RTDW) framework based on the virtualization concept. Our approach introduces a conceptual modelling technique, known as ring modelling, for real‐time data management and multidimensional analysis. This technique produces a flexible semi‐structured data model that accommodates unknown business process data and relationships as they evolve, handles schema changes and aggregate‐management efficiently, and scales well with the large size of increasing data volumes. With the help of a telecommunication business example, We evaluated our proposed approach in an extensive experimental study where we compared our approach Ring Model with existing structured multidimensional conceptual models (MCMs), i.e. relational OLAP and multidimensional OLAP, and with semi‐structured MCM, i.e. XML Cubes, in terms of scalability, data storage estimations, data updates loading time, and query response times. Our performance results show that encouraging speedups are achieved.

show abstract

Section: Storage Estimations and Incremental Loading Of Data Updatesmentioning

confidence: 99%

Section: Dwvf and Prototype Implementationmentioning

confidence: 99%